Improved flavodoxin inhibitors with potential therapeutic effects against Helicobacter pylori infection

Article abstract of DOI:10.1021/jm400786q

Helicobacter pylori (Hp) infection affects one-half of the human population and produces a variety of diseases from peptic ulcer to cancer. Current eradication therapies achieve modest success rates (around 70%), resistance to the antibiotics of choice is on the rise, and vaccination has not proved to be successful yet. Using an essential Hp protein, flavodoxin, as target, we identified three low-molecular-weight flavodoxin inhibitors with bactericidal anti-Hp properties. To improve their therapeutic indexes, we have now identified and tested 123 related compounds. We have first tested similar compounds available. Then we have designed, synthesized, and tested novel variants for affinity to flavodoxin, MIC for Hp, cytotoxicity, and bactericidal effect. Some are novel bactericidal inhibitors with therapeutic indexes of 9, 38 and 12, significantly higher than those of their corresponding leads. Developing novel Hp-specific antibiotics will help fighting Hp resistance and may have the advantage of not generally perturbing the bacterial flora.

Full text of DOI:10.1021/jm400786q

Article
pubs.acs.org/jmc
Improved Flavodoxin Inhibitors with Potential Therapeutic Effects
against Helicobacter pylori Infection
Juan J. Galano,†,‡ Miriam Alías,†,‡ Reyes Perez,†,‡ Adrian Velazquez-Campoy,†,‡,§ Paul S. Hoffman,∥
́
́
and Javier Sancho*,†,‡
†Departamento de Bioquímica y Biología Molecular y Celular, Facultad de Ciencias, Universidad de Zaragoza, 50009, Zaragoza, Spain
‡Institute for Biocomputation and Physics of Complex Systems (BIFI), Joint Unit BIFI-IQFR (CSIC), Edificio I + D, Mariano
Esquillor, 50018, Zaragoza, Spain
§Fundacion
́
ARAID, Gobierno de Aragon, Aragon, Spain
∥Department of Medicine, Division of Infectious Diseases and International Health, University of Virginia Health System,
Charlottesville, Virginia 22908, United States
S
* Supporting Information
ABSTRACT: Helicobacter pylori (Hp) infection affects one-
half of the human population and produces a variety of
diseases from peptic ulcer to cancer. Current eradication
therapies achieve modest success rates (around 70%),
resistance to the antibiotics of choice is on the rise, and
vaccination has not proved to be successful yet. Using an
essential Hp protein, flavodoxin, as target, we identified three
low-molecular-weight flavodoxin inhibitors with bactericidal anti-Hp properties. To improve their therapeutic indexes, we have
now identified and tested 123 related compounds. We have first tested similar compounds available. Then we have designed,
synthesized, and tested novel variants for affinity to flavodoxin, MIC for Hp, cytotoxicity, and bactericidal effect. Some are novel
bactericidal inhibitors with therapeutic indexes of 9, 38 and 12, significantly higher than those of their corresponding leads.
Developing novel Hp-specific antibiotics will help fighting Hp resistance and may have the advantage of not generally perturbing
the bacterial flora.
metronidazole. Unfortunately, this combination has lost efficacy
and, at present, it allows the cure of a maximum of 70% of the
patients. The reason for this loss of efficacy appears to be the
increase in clarithromycin and/or metronidazole resistance in
Hp. Besides, in developing countries eradication therapies face
the additional problems of high cost, high prevalence of strains
resistant to available antibiotics,6,7 and/or high reinfection rates
due to poor socioeconomic and sanitation policies. As
indicated, “a wider range of effective treatments is urgently
required.”3 In contrast with needs, no new drugs have been
developed for this indication in recent years and, significantly
enough, there is not a single specific Hp antimicrobial.
Novel molecules, synthetic or natural, showing anti-Hp
activity have been described elsewhere (see ref 8 for a recent
review). Often, such novel compounds are not identified as
active against specific Hp validated targets, which complicates
their further development, and so far, they have not resulted in
novel antibiotics. Vaccination is a clear alternative to combat
Hp infection, but although there have been many initiatives to
obtain Hp vaccines, no one has succeeded yet. A recent review
on this approach9 states that “... vaccine development against H.
pylori remains a focus of research. Progress is made but is
INTRODUCTION
■
Helicobacter pylori (Hp) is a Gram negative bacteria that
establishes life-long infections in the gastric mucosa of
humans.1,2 In many cases, without specific antimicrobial
intervention, Hp infected individuals will develop type B
gastritis, chronic peptic ulcers, and more rarely, gastric
neoplasias. As stated in the Maastricht IV/Florence Consensus
Report,3 “H. pylori is the most successful human pathogen,
infecting an estimated 50% of the global population.” The
prevalence of the infection varies from place to place, average
prevalence in Europe being around 30% and much higher in
developing countries. Individuals infected with Hp have a 10−
20% lifetime risk of developing peptic ulcers and a 1−2% risk of
developing stomach cancer. Eradication of Hp infection is
recommended in peptic ulcer disease, low grade gastric mucosa
associated lymphoid tissue (MALT) lymphoma, atrophic
gastritis, first degree relatives of patients with gastric cancer,
unexplained iron deficiency anemia, and chronic idiopathic
thrombocytopenic purpura. Besides, it has been pointed out
that Hp eradication may prevent peptic ulcer in naive users of
nonsteroidal anti-inflammatory drugs (NSAIDs).4
The standard therapy used worldwide to eradicate Hp,
known as the triple treatment,5 consists of a combination of
one proton pump inhibitor (PPI) with two wide spectrum
antibiotics: clarithromycin and a choice of amoxicillin or
Received: May 19, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Helicobacter pylori flavodoxin X-ray structure. (A) Ribbon drawing showing the structure of Hp-Fld and a transparent molecular surface of
the cofactor binding site (the FMN cofactor in sticks). (B) Molecular surface of the cofactor binding site showing the FMN-interacting Y92 residue
and A55. Hp-Fld A55 appears at a position where bulkier residues are typically located in other flavodoxins, thereby helping to create a pocket near
the active site where it is believed the inhibitors to bind.17
incremental. There is need for a still better understanding of
the protective mechanism and for improving efficacy.” Given
that the existing anti-Hp products (wide spectrum antibiotics
typically used in combinations) currently achieve only a 70%
success in eradication, novel, ideally Hp-specific small
molecules are strongly needed to face the challenge of growing
resistance. On the other hand, development of antibiotics
specific to a given bacteria may be advantageous in that it may
reduce the characteristic side effects of antibiotic therapy
consisting of disruption of the natural human bacterial flora.10
Several groups have attempted to develop narrow-spectrum
antimicrobials against Hp and have indeed identified selective
targets in this organism.11 One of them is the protein
flavodoxin (Hp-Fld), whose function is essential for Hp
viability.12,13 Flavodoxins are electron carriers involved in a
wide range of bacterial reactions, and they contain a redox
active flavin mononucleotide molecule (FMN) noncovalently
bound.14 Hp-Fld is involved in the oxidative decarboxylation of
pyruvate leading to synthesis of NADPH. In this reaction, Hp-
Fld shuttles electrons from the pyruvate−ferredoxin oxido-
reductase complex (PFOR)15 to flavodoxin−quinone reductase
(FqrB).16 Structurally, Hp-Fld differs from other flavodoxins in
that it forms a distinct pocket near the cofactor binding site
(Figure 1) where an alanine residue replaces the structurally
equivalent bulky residue typically present in other flavodox-
ins.13 We proposed12 that the pocket could allow a selective
binding of inhibitors that could interfere with flavodoxin
function and accordingly screened a 10 000-compound
collection using an in vitro protein thermostability assay.17
Thus, we identified 29 compounds that specifically bind to Hp-
Fld and could in principle inhibit its function. The 29 binders
were tested in a flavodoxin functional assay performed in
anaerobiosis in the presence of the PFOR and FqrB enzymes,
which allowed us to identify four flavodoxin inhibitors. Those
four compounds proved to inhibit the growth of Hp cultures,
and three of them showed bactericidal effects.17
INHIBITOR CANDIDATES
■
The compounds tested include three previously reported
flavodoxin inhibitors,17 compounds I, II, and IV, plus 123
related compounds: 30 analogues of I, 62 analogues of II, and
31 analogues of IV. Analogues of I included 22 compounds
identified by similarity searches in the Maybridge catalogue
(http://www.maybridge.com) (Supporting Information Table
S1, compounds 1−4, 8−25), 2 compounds identified by
similarity searches in SciFinder database (http://scifinder.cas.
com) (Table S1, compounds 5, 26), and 6 compounds
purposely designed that were synthesized by Maybridge (Table
S1, compounds 6, 7, 27−30). Analogues of II included 44
compounds identified by similarity searches in the Maybridge
catalogue (Table S1, compounds 31, 32, 46−87), 10
compounds identified by similarity searches in Scifinder
database (Table S1, compounds 33−39, 88−90), and 8
compounds purposely designed, which were synthesized by
Maybridge (Table S1, compounds 40−45, 91, 92). Analogues
of IV included 19 compounds identified by similarity searches
in the Maybridge catalogue (Table S1, compounds 93−110,
122), 5 compounds indentified by similarity searches in
Scifinder (Table S1, compounds 111−115), and 7 compounds
purposely designed, which were synthesized by Maybridge
(Table S1, compounds 116−121, 123). The 17 compounds
identified in the SciFinder database were either purchased from
commercial sources or kindly supplied by a variety of
laboratories (see suppliers in Table S1).
SYNTHESIS
■
Briefly, the custom-made syntheses performed by Maybridge
Chemical Company, Ltd. leading to the new analogues were as
follows.
Schemes 1 and 2 show the synthetic routes to new analogues
of II. Nitrostyrene derivatives 40−45 and 91 were accessible via
condensation of nitromethane with appropriate aldehydes (see
Table S1 for identities of R1−R5) under either base catalyzed
(method A, for 40−42 and 44)18 or acid catalyzed conditions
(method B, for 43, 45, and 91).19
Compound 92 could be readily prepared via Horner−
Wadsworth−Emmons olefination20 of the appropriate aldehyde
131.
Analogues of I were prepared according to Schemes 3 and 4.
Some of the already synthesized nitrostyrene derivatives (43−
We describe here the identification/design and testing of 123
compounds related to those three bactericidal compounds,
which have led to the discovery of novel bactericidal
compounds with significantly greater therapeutic indexes.
B
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
derivatives of I, II, and IV, which were available from
Maybridge and those identified in SciFinder and commercially
available or gifted to us by academic laboratories were initially
tested for binding to Hp-Fld using a fast thermostabilization
assay implemented in the previously reported HTS.17 The
compounds that increased the Hp-Fld Tm by at least 2 standard
deviations of the mean Tm obtained for the control samples
(with no compound added) were selected for further testing.
They include analogues of I (1−5), analogues of II (31−38),
and analogues of IV (93, 102, 105−108, and 110). The affinity
of these compounds for Hp-Fld was then determined by ITC.
The analogues of I selected from their thermostabilization of
flavodoxin were shown by ITC to form 1:1 complexes with the
protein with Kd from 0.4 to 1.9 μM, in the same range that the
Kd of the complex formed between flavodoxin and I: 1.0 μM
(Table 1). The selected analogues of II formed 1:1 complexes
with flavodoxin with Kd covering a wider range of values (from
0.7 to 19 μM) around the value measured for the complex
between flavodoxin and II: 3.5 μM. The selected analogues of
IV also established 1:1 complexes with flavodoxin with Kd from
0.2 to 9.6 μM, around the Kd value for the complex between
flavodoxin and IV: 1.7 μM.
MICs and Bactericidal Properties of Derivatives. The
selected Hp-Fld binders were tested as inhibitors of Hp growth
by determining their MICs. Out of the 20 binders identified, 19
were effective inhibiting Hp growth, with MICs ranging from
1.2 to 7.5 μM for analogues of I, from 2.4 to 8 μM for
analogues of II, and from 4.8 to 38 μM for analogues of IV
(Table 1). Kill curves determined for some of the Hp inhibitors
showing the highest therapeutic indexes (computed using the
MICs and the toxicity data (MCC) described below) indicated
that at least two analogues of I (4 and 5), one analogue of II
(31), and two analogues of IV (105 and 108) displayed
bactericidal properties.
Toxicity of Derivatives and Therapeutic Indexes. The
toxicity for HeLa cells of the 19 binders that inhibit Hp growth
was determined using the XTT assay. The corresponding
MCCs are reported in Table 1. They range from 0.01 μM for
the most toxic compound to 100 μM for several less toxic
compounds. With the exception of three derivatives of II, all
the MCCs were ≥1 μM. The TIs corresponding to derivatives
of I ranged from 0.4 to 9.3, those for derivatives of II, from
0.001 to 8.4, and those for derivatives of IV, from 0.2 to 11.
Design and Testing of Novel Analogues. The thermo-
stabilization (not shown) and the binding data obtained using
the available analogues of I, II, and IV were qualitatively
analyzed in order to design novel variants. For compound I,
substitution of chlorine at position 6 in the chromene moiety
appeared preferred to substitution of chlorine at position 8,
while in the fluorophenyl moiety, fluorosubstitution was
preferred to introduction of substituents at other positions
(not shown). Variants were designed to test the importance of
the nitro group for binding and activity (27 and 7), to test
Scheme 1
Scheme 2
45; see Table S1 for identities of R1) were further used to
obtain compounds 6 and 28−30, via Baylis−Hillman reaction21
with the appropriate aldehydes (132−134; see Table S1 for
identities of R2−R5 and X).
The synthesis of compounds 7 and 27 is described in
Scheme 4. The first step consists of a Vilsmeier−Haack
reaction22 on 135, followed by oxidation23 of the resulting
aldehyde 136. Further conversion of the carboxylic acid 137 to
the ethyl ester 138 was achieved in a one-pot reaction.24 This
was followed by Grignard reaction25 of 138 with 1-bromo-4-
fluorobenzene, which gave the desired compound 139 as a
mixture of keto−enol tautomers (139a and 139b). Subsequent
reduction of the mixture with sodium borohydride26 afforded
the alcohol 140 and also some of the desired ester 7. This crude
mixture of alcohol and ester was dehydrated with p-
toluensulfonic acid,27 leading to compound 7, which was
quantitatively hydrolyzed under basic conditions28 to obtain
compound 27.
The synthesis of analogues of IV is described in Scheme 5.
Compounds 116−119, 121, and 123 could be obtained
applying the same strategy, consisting of a reaction of the
appropriate thiols with the appropriate chlorobenzoxadiazoles
under basic conditions.29 The thiols required for the synthesis
of 117 and 118 were 142 and 144, respectively, which were
prepared from 141 and 143 by reaction with thiourea and
subsequent hydrolysis.30 Thiol 146, which was used for the
synthesis of 123, was obtained from 145 by reaction with 2-
aminoethanothiol.31 Also, one of the intermediates in the
synthesis of 119 was further used for the preparation of 121.32
Finally, 120 was obtained by reduction33 of the convenient
commercially available nitrobenzoxadiazole 150.
RESULTS
■
Prescreening of Available Analogues for Binding to
Hp Flavodoxin and Affinities of the Complexes. The
Scheme 3
C
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Scheme 4
Scheme 5
different halogen replacements for the fluorine (29 and 6), to
test replacement of the chromene oxygen atom by nitrogen
(28), and to introduce substituents at other positions in
chromene ring (30) (Table 2 and Table S1). Replacement of
the nitro group by a carboxylic acid (27) lowered the affinity
for flavodoxin, and formation of complex was not observed by
D
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. Observed Values for Tested BRs in Selected Compounds
E
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. continued
F
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. continued
G
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. continued
a
b
c
d
e
Number of compound given in Table S1. Values calculated as MCC/MIC. Derivative of inhibitor I. Derivative of inhibitor II . Derivative of
inhibitor IV.
ITC (not shown). Esterification (7) restored binding with Kd
similar to that of I. The toxicity of 7 was slightly lower than that
of I, but unfortunately, the MIC was greatly increased, leading
to a lower TI compared to I. Introduction of a chlorine at
position 5 (30) led to loss of binding (not shown).
Replacement of the fluorine by either bromine or iodine (29,
6) or replacement of the oxygen by nitrogen (28) lowered the
affinity by 1 order of magnitude (Kd for 6 of 25 μM (Table 1);
Kd for 28 and 29 of 33 and 25 μM, respectively). The iodine
containing derivative (6) nevertheless displayed similar MIC,
toxicity, and TI compared with I.
For compound II, the preliminary thermostabilization and
the affinity data suggested that the trifluoromethyl groups could
be removed or replaced and that alkylation of the vinyl moiety
was possible. Replacement of the nitro group by a carboxylic
acid (39) reduced the affinity, but complex formation with
flavodoxin was still observed. Although 39 was less toxic than
II, it was also less effective (higher MIC) and the TI was lower.
Esterification (92) (see Table S1) greatly reduced the affinity,
and no complex formation was observed by ITC (not shown).
Removal of one CF3 group and insertion of a chlorine at the
position para to the nitrovinyl substituent (40) hardly changed
affinity, inhibition, or toxicity, and the compound exhibited a TI
similar to that of II.
Replacement of both CF3 groups by one chlorine and one
bromine (91) (see Table S1) left the affinity and toxicity almost
unchanged (Kd of 10 μM and MCC of 2.7 μM). In compound
41 one of the CF3 groups was removed and fluorine was
introduced in the position para to the remaining CF3. This
compound displayed similar affinity, lower toxicity, and lower
MIC, thus displaying a higher TI (∼15) than II. Compound 42
contained no CF3 groups but two fluorines (both ortho to the
nitrovinyl substituent) plus one chlorine (para to the nitrovinyl
substituent). This compound exhibited a slightly lower affinity
but was more potent (lower MIC) and 1 order of magnitude
less toxic than II. Its TI (∼38) was thus higher. Finally, three
compounds were tested that contained no CF3 groups but just
one halogen (in para position): fluorine, bromine, or iodine
(45, 43, or 44, respectively). Only the fluorine containing
compound displayed a clearly lower affinity than II. The three
H
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
compounds were as inhibitory or more inhibitory than II, and
they exhibited lower toxicity. Accordingly their TIs were higher
than those of II.
Similarly, for compound II such replacements of the nitro
group reduced both the inhibitory activity and the cytotoxicity.
Interestingly, the simultaneous removal of the two CF3 groups
with introduction of halogens in the para position markedly
reduced the toxicity, greatly increasing the TI.
The preliminary analysis of derivatives of IV indicated that
the p-methoxyphenyl moiety could be modified and that the
spacing between this moiety and the sulfur atom could be
changed. The relevance of the spacing was tested by directly
connecting the p-methoxyphenyl group to the sulfur atom
(116) and by replacing the methylene group in IV by either an
ethylene (117) or a propylene group (118). Only compound
117 (see Table S1) showed no affinity for flavodoxin (not
shown), 116 displaying a similar Kd and 118 being significantly
less affine but nevertheless binding to the protein. The spacing
was also modified by introducing an amide linkage (compound
123; see Table S1) between the two aromatic systems in IV,
but this compound did not apparently bind to flavodoxin (not
shown). Of these four derivatives, 116 exhibited the highest TI
(=1.5), which, although admittedly small, is higher than that of
IV (∼0.2).
Replacement of the nitro group by a sulfonamide group
(119) improved the affinity, lowered the toxicity, and slightly
improved the inhibition, leading to an improved TI of 0.6.
Replacement of the nitro group by an amine group (120)
decreased the affinity, did not modify the inhibition, and
markedly reduced the toxicity, which combined to a much
higher TI of 12. Two short versions of IV were additionally
tested. In one of them (121), the nitro group, was replaced by a
sulfonamide group, and the p-methoxyphenyl moiety was
replaced by a hydroxymethyl group. This compound can still
bind flavodoxin (with lower affinity) and displays a low toxicity,
but it cannot inhibit the growth of Hp in culture. In the other
one, only the benzoxadiazole ring remains, with the oxygen
changed by sulfur and one bromine as the only substituent
(122). Its properties are very similar to those of the other
shortened compound, and it does not inhibit Hp cells growth.
The bactericidal properties of some of the novel compounds
exhibiting high TI have been tested. Compounds 42−45, 116,
and 120 are bactericidal.
On the other hand, inhibitor IV contains two moieties
(benzoxadiazole and p-methoxyphenyl) joined by a thioether
link. The influence of link length was tested, and it seems that
shorter links (with no or only one methylene) are better than
longer ones. On the other hand, the nitro group could be
replaced by a sulfonamide group with some improvement of
the TI. More importantly, replacement of the nitro group by an
amine markedly reduced the toxicity and therefore improved
the TI. As for the p-methoxyphenyl moiety, two attempts to
essentially remove it led to smaller molecules that could bind
flavodoxin and displayed reduced toxicity but lacked inhibitory
activity. Both of these variants lacked the nitro group.
The variants of I, II, and IV exhibiting the highest TI within
their groups were tested for bactericidal activity and they did
kill Hp cells. They thus constitute improved versions of
inhibitors I, II, and IV with significantly expanded therapeutic
windows.
CONCLUSIONS
■
Three known bactericidal compounds potentially useful to fight
Hp infection have been improved using a two-round testing of
derivatives. Initially, existing variants have been evaluated and,
based on the results, novel variants have been designed and
tested for affinity, toxicity, and inhibition of Hp growth. Overall,
the therapeutic index of compound I has been increased from 5
to 9, that of II from 1.5 to 38, and that of IV from 0.2 to 12.
The derivatives of each compound exhibiting the highest TI are
bactericidal and will be subjected to efficacy testing in an animal
model to evaluate their feasibility as leads for novel specific
antibiotics for the treatment of Helicobacter pylori associated
diseases.
EXPERIMENTAL SECTION
Reagents and Chemicals. All reagents and chemicals were
■
DISCUSSION
■
obtained from commercial suppliers and were used without any further
An initial test of the binding of existing derivatives of inhibitors
I, II, and IV to flavodoxin was performed using a HTS
method.17 The test allowed us to identify five (1−5), eight
(31−38), and seven (93, 102, 105−108, and 110) novel
binders structurally related to, respectively, compounds I, II,
and IV (Table 1).
1
purification. H NMR and 19F NMR spectra were acquired at room
temperature at 400 and 376 MHz, respectively, using a 5 mm probe.
The chemical shifts (δ) are reported in parts per million from
tetramethylsilane with the solvent resonance as the internal standard.
Coupling constants (J) are quoted in hertz. The splitting patterns are
reported as s (singlet), d (doublet), t (triplet), q (quartet), dd (doublet
of doublets), m (multiplet), bs (broad singlet). Purity of compounds
(made to order or kindly supplied by other laboratories) was
determined to be ≥95% by HPLC chromatography, using a Waters
HPLC system equipped with a 600-E pump, a 2996 PDA detector, and
a 2707 autosampler. The LC system was fitted with a C18 reversed-
phase column (VYDAC 238TP C18 5 μm, 4.6 mm × 250 mm) and
operated, unless otherwise stated, using a linear gradient of buffer B
(100% in 40 min) from 100% buffer A (buffer A consisting of 0.1%
TFA in H2O, buffer B consisting of 0.085% TFA in CH3CN/H2O,
95:5 v/v) at a flow rate of 1 mL/min.
Hp Flavodoxin. The flavodoxin from H. pylori was expressed in E.
coli and purified as reported.12 Protein concentration was determined
from the absorbance of the FMN cofactor at 452 nm using an
extinction coefficient of 10 650 M−1 cm−1.12
Preliminary Testing for Binding to Hp Flavodoxin. For some
of the inhibitor candidates, a preliminary flavodoxin binding test was
performed in order to early discard compounds not binding to the
target protein. This screening was based on comparing the
temperature of mid-denaturation (Tm) of flavodoxin with that of
In each of the three groups, there were compounds binding
to flavodoxin with higher affinity than the original inhibitors.
Most of these binders (19 out of 20) were effective at inhibiting
Hp growth, and in each of the three groups there were
bactericidal compounds. Interestingly, some of these inhibitors
exhibited a significantly lower toxicity toward HeLa cells, which
led to an increase in their TI. On the basis of the data gathered
for these compounds, in some cases suggesting which parts of
the molecules could be modified without compromising affinity
for the flavodoxin target, we designed novel variants of
inhibitors I, II, and IV, aiming at exploring their chemical
space while trying to introduce single modifications whenever
possible and taking into account synthetic issues.
Analysis of activity data from the novel derivatives of I
indicated that its nitro group could not be replaced by either a
carboxylic acid or the corresponding ethyl ester without
severely compromising either binding or inhibitory efficiency.
I
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
flavodoxin in the presence of compound.17 To that end, the intrinsic
fluorescence of flavodoxin in the visible was monitored with a FluoDia
T70 spectrofluorimeter (excitation at 445 nm and emission at 525 nm)
as a function of temperature. The flavodoxin FMN cofactor, strongly
quenched by the apoprotein, is released as flavodoxin unfolds giving
rise to a large increase in fluorescence emission.34 The thermal
unfolding curves were analyzed using software developed by us, and
the apparent Tm values were calculated as reported.17 Only the
compounds that increased the Tm of the protein by at least 2 standard
deviations of the mean Tm of protein controls (with no compound
added) were further tested.
Determination of the Affinity of Complexes between Hp-Fld
and Candidate Inhibitors. The dissociation constants of the
complexes were determined at 25 °C by isothermal titration
calorimetry (ITC) using a VP-ITC titration calorimeter (MicroCal,
GE Healthcare). Degassed 20 μM flavodoxin solutions were titrated
with concentrated inhibitor solutions (around 500 μM) dissolved in
the same buffer (50 mM EPPS, pH 9.0, and 5% DMSO), and the heats
associated with each injection were fitted assuming a 1:1 stoichiometry
for the complex formed.
Minimal Inhibitory Concentrations (MICs). Hp strains 26695
and Hp1061 were grown as described.16 For microdilution MIC
determinations, 96-well round-bottom microtiter dishes were used.
Bacteria growth and subsequent dilution to OD660nm of 0.01 prior to
mixing with appropriate concentrations of test compound were
performed as described.17 The low percentage of DMSO present in
the assay (<3% v/v) was not deleterious for the cells. Dishes were
shaken under microaerobic conditions for 28 h. The lowest compound
concentration completely inhibiting Hp growth was recorded as the
MIC of the compound.
Bactericidal Assays. The bactericidal activity of compounds
toward Hp cells (strains 26695 and Hp1061) growing in Brucella-based
medium supplemented with 7.5% fetal bovine serum was determined
as previously described.16 Bacteria were diluted to OD660nm = 0.1 in 3
mL of BHI broth supplemented with 2% newborn calf serum, and the
appropriate small volume of test compound was added to a final test
concentration of 2 × MIC. Aliquots of the culture were removed at
different time points, centrifuged, resuspended in BHI broth, and
plated on Brucella agar supplemented with 7.5% serum for
determination of viable counts. Bactericidal activity was calculated
following comparison with the DMSO control.
Minimal Cytotoxic Concentrations (MCCs). The toxicity of the
compounds toward HeLa cells was determined by the XTT method
using the Cell Proliferation Kit II (Roche), which detects dehydrogen-
ase activity by reduction of a tetrazolium salt. All experiments were
performed twice in triplicate. HeLa cells were cultured in complete
medium (500 mL of Dulbecco’s modified Eagle medium: P04-03591
from Ibian Technologies plus 50 mL of bovine fetal serum plus 5.5 mL
of antibiotic containing 550 000 units of penicillin and 550 000 μg of
streptomycin) with phenol red using 96-well plates (with 30 000 cells
in 100 μL in each well) to which 1 μL volumes of compound dissolved
in DMSO were added to final compound concentrations of 0.1, 0.25,
0.5, 1, 10, 25, 50, 75, and 100 μM. To control wells, 1 μL of DMSO
was added. Plates were incubated at 37 °C for 24 h. Then they were
centrifuged. The medium was replaced by 100 μL of fresh medium
without phenol red, and 50 μL of XTT-PMS mixture (50 μL of XTT +
1 μL of PMS) was added to each well. The plates were incubated for 4
h at 37 °C, and the absorbance at 450 and 500 nm was recorded in an
ELISA reader. Cell viability was calculated as explained.35
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: jsancho@unizar.es. Phone: (+34) 976 761286. Fax:
(+34) 976 762123.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
J.J.G is a recipient of a Ph.D. studies fellowship awarded by
Banco Santander and University of Zaragoza (Spain). This
work was supported by Grants BFU2010-16297 and BFU2010-
19451 (Ministerio de Ciencia e Innovacion
and Grupo Protein Targets B89 (Diputacion
Aragon, Spain).
́
, MICINN, Spain)
́
General de
́
ABBREVIATIONS USED
■
EPPS, 4-(2-hydroxyethyl)-1-piperazinepropanesulfonic acid or
4-(2-hydroxyethyl)piperazine-1-propanesulfonic acid; DMF,
dimethylformamide; DCM, dichloromethane; XTT, tetrazo-
lium salt; PMS, N-methylphenazonium methyl sulfate; DMSO,
dimethyl sulfoxide; BHI, brain−heart infusion; VP-ITC,
pressure and volume constant isothermal titration calorimetry;
OD, optical density; TMS, tetramethylsilane; TFA, trifluoro-
acetic acid; p-TSA, p-toluensulfonic acid; BR, biological
response; TI, therapeutic index
REFERENCES
■
(1) Bruce, M. G.; Maaroos, H. I. Epidemiology of Helicobacter pylori
infection. Helicobacter 2008, 13, 1−6.
(2) Marshall, B. J.; Warren, J. R. Unidentified curved bacilli in the
stomach of patients with gastritis and peptic ulceration. Lancet 1984, 1,
1311−1315.
(3) Malfertheiner, P.; Megraud, F.; O’Morain, C. A.; Atherton, J.;
Axon, A. T.; Bazzoli, F.; Gensini, G. F.; Gisbert, J. P.; Graham, D. Y.;
Rokkas, T.; El-Omar, E. M.; Kuipers, E. J. Management of Helicobacter
pylori infectionthe Maastricht IV/Florence consensus report. Gut
2012, 61, 646−664.
(4) Malfertheiner, P.; Megraud, F.; O’Morain, C.; Bazzoli, F.; El-
Omar, E.; Graham, D.; Hunt, R.; Rokkas, T.; Vakil, N.; Kuipers, E. J.
Current concepts in the management of Helicobacter pylori infection:
the Maastricht III consensus report. Gut 2007, 56, 772−781.
(5) Malfertheiner, P.; Megraud, F.; O’Morain, C.; Bell, D.; Bianchi
Porro, G.; Deltenre, M.; Forman, D.; Gasbarrini, G.; Jaup, B.;
Misiewicz, J. J.; Pajares, J.; Quina, M.; Rauws, E.; The European
Helicobacter Pylori Study Group (EHPSG).. Current European
concepts in the management of Helicobacter pylori infectionthe
Maastricht consensus report. Eur. J. Gastroenterol. Hepatol. 1997, 9, 1−
2.
(6) Martinez-Julvez, M.; Rojas, A. L.; Olekhnovich, I.; Espinosa
Angarica, V.; Hoffman, P. S.; Sancho, J. Structure of RdxAan
oxygen-insensitive nitroreductase essential for metronidazole activa-
tion in Helicobacter pylori. FEBS J. 2012, 279, 4306−4317.
(7) Secka, O.; Berg, D. E.; Antonio, M.; Corrah, T.; Tapgun, M.;
Walton, R.; Thomas, V.; Galano, J. J.; Sancho, J.; Adegbola, R. A.;
Thomas, J. E. Antimicrobial susceptibility and resistance patterns
among Helicobacter pylori strains from The Gambia, West Africa.
Antimicrob. Agents Chemother. 2013, 57, 1231−1237.
ASSOCIATED CONTENT
■
S
* Supporting Information
Suppliers and molecular structures of all inhibitors tested and
NMR data. This material is available free of charge via the
Internet at http://pubs.acs.org.
(8) Fiorini, G.; Zullo, A.; Gatta, L.; Castelli, V.; Ricci, C.; Cassol, F.;
Vaira, D. Newer agents for Helicobacter pylori eradication. Clin. Exp.
Gastroenterol. 2012, 5, 109−112.
J
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(9) Aebischer, T.; Meyer, T. F.; Andersen, L. P. Inflammation,
immunity, and vaccines for Helicobacter. Helicobacter 2010, 15
(Suppl.1), 21−28.
(32) Choi, J. S.; Lee, H.-S.; Lee, Y.; Jeong, N.; Kim, H.-J.; Kim, Y.-D.;
Han, H. Nsc-mediated solid-phase synthesis of polyamides containing
pyrrole amino acid. Tetrahedron Lett. 2002, 43, 4295−4299.
(33) Hazlet, S. E.; Dornfeld, C. A. The reduction of aromatic nitro
compounds with activated iron. J. Am. Chem. Soc. 1944, 66, 1781−
1782.
(34) Cremades, N.; Velazquez-Campoy, A.; Freire, E.; Sancho, J. The
flavodoxin from Helicobacter pylori: structural determinants of
thermostability and FMN cofactor binding. Biochemistry 2008, 47,
627−639.
(35) Lopez, L. C.; Dos-Reis, S.; Espargaro, A.; Carrodeguas, J. A.;
Maddelein, M. L.; Ventura, S.; Sancho, J. Discovery of novel inhibitors
of amyloid beta-peptide 1−42 aggregation. J. Med. Chem. 2012, 55,
9521−9530.
(10) Blaser, M. Antibiotic overuse: stop the killing of beneficial
bacteria. Nature 2011, 476, 393−394.
(11) Chalker, A. F.; Minehart, H. W.; Hughes, N. J.; Koretke, K. K.;
Lonetto, M. A.; Brinkman, K. K.; Warren, P. V.; Lupas, A.; Stanhope,
M. J.; Brown, J. R.; Hoffman, P. S. Systematic identification of selective
essential genes in Helicobacter pylori by genome prioritization and
allelic replacement mutagenesis. J. Bacteriol. 2001, 183, 1259−1268.
(12) Cremades, N.; Bueno, M.; Toja, M.; Sancho, J. Towards a new
therapeutic target: Helicobacter pylori flavodoxin. Biophys. Chem. 2005,
115, 267−276.
(13) Freigang, J.; Diederichs, K.; Schafer, K. P.; Welte, W.; Paul, R.
Crystal structure of oxidized flavodoxin, an essential protein in
Helicobacter pylori. Protein Sci. 2002, 11, 253−261.
(14) Sancho, J. Flavodoxins: sequence, folding, binding, function and
beyond. Cell. Mol. Life Sci. 2006, 63, 855−864.
(15) Hughes, N. J.; Chalk, P. A.; Clayton, C. L.; Kelly, D. J.
Identification of carboxylation enzymes and characterization of a novel
four-subunit pyruvate:flavodoxin oxidoreductase from Helicobacter
pylori. J. Bacteriol. 1995, 177, 3953−3959.
(16) St Maurice, M.; Cremades, N.; Croxen, M. A.; Sisson, G.;
Sancho, J.; Hoffman, P. S. Flavodoxin:quinone reductase (FqrB): a
redox partner of pyruvate:ferredoxin oxidoreductase that reversibly
couples pyruvate oxidation to NADPH production in Helicobacter
pylori and Campylobacter jejuni. J. Bacteriol. 2007, 189, 4764−4773.
(17) Cremades, N.; Velazquez-Campoy, A.; Martinez-Julvez, M.;
Neira, J. L.; Perez-Dorado, I.; Hermoso, J.; Jimenez, P.; Lanas, A.;
Hoffman, P. S.; Sancho, J. Discovery of specific flavodoxin inhibitors as
potential therapeutic agents against Helicobacter pylori infection. ACS
Chem. Biol. 2009, 4, 928−938.
(18) Kumaran, G.; Gurunath, H. K. Titanium (IV) chloride−
triethylsilane mediated conversion of ω-nitrostyrenes to phenyl-
acetohydroximoyl chlorides. Tetrahedron Lett. 1994, 35, 9099−9100.
(19) Ju-Tsung, L.; Ching-Fa, Y. One-pot synthesis of trans-β-
alkylstyrenes. Tetrahedron Lett. 2001, 42, 6147−6150.
(20) Wadsworth, W. S. Synthetic applications of phosphoryl-
stabilized anions. Org. React. 2005, 73−253.
(21) Basavaiah, D.; Rao, A. J.; Satyanarayana, T. Recent advances in
the Baylis−Hillman reaction and applications. Chem. Rev. 2003, 103,
811−892.
(22) Marson, C. M. Reactions of carbonyl compounds with
(monohalo)methyleniminium salts (Vilsmeier reagents). Tetrahedron
1992, 48, 3959−3726.
(23) Balkrisna, S. B.; Childers, W. E. J.; Pinnick, H. W. Oxidation of
α,β-unsaturated aldehydes. Tetrahedron 1981, 37, 2091−2096.
(24) Verma, R. K.; Kumar, B. Esterification at room temperature: a
mixing affair only. Synth. Commun. 1984, 14, 1359−1363.
(25) Posner, G. H. Conjugate addition reactions of organocopper
reagents. Org. React. 1972, 19, 1−113.
(26) Chaikin, S. W.; Brown, W. G. Reduction of aldehydes, ketones
and acid chlorides by sodium borohydride. J. Am. Chem. Soc. 1949, 71,
122−125.
(27) Larock, R. Comprehensive Organic Transformations; VCH: New
York, 1989; pp 151−152.
(28) Larock, R. Comprehensive Organic Transformations. VCH: New
York, 1989; pp 981−985.
(29) Cosimelli, B.; Roncucci, G.; Dei, D.; Fantetti, L.; Ferroni, F.;
Ricci, M.; Spinelli, D. Synthesis and antimycotic activity of new
unsymmetrical substituted zinc phthalocyanines. Tetrahedron 2003, 59,
10025−10030.
(30) Urquhart, G. G., Gates, J. W.; Connor, R. Organic Syntheses.;
Wiley & Sons: New York, 1955; Collect. Vol. 3, pp 363−365.
(31) Petersson, M. J.; Jenkins, I. D.; Loughlin, W. A. The use of
phosphonium anhydrides for the synthesis of 2-oxazolines, 2-
thiazolines and 2-dihydrooxazine under mild conditions. Org. Biomol.
Chem. 2009, 7, 739−746.
K
dx.doi.org/10.1021/jm400786q | J. Med. Chem. XXXX, XXX, XXX−XXX